Unstable angina is a frequent diagnosis in post-menopausal women and is associated with a significant risk of myocardial infarction and need for revascularization. The pathogenesis of unstable angina involves vasoconstriction superimposed on fixed disease causing a temporary decrease in coronary blood flow. Recent catheterization studies in patients with atherosclerosis utilizing quantitative angiography and intracoronary doppler measurements of blood flow velocity suggest that endothelial dysfunction results in a paradoxical coronary vasoconstriction response to certain neurohumoral stimuli including acetylcholine, catecholamines and serotonin and resultant myocardial ischemia. Therapeutic agents which prevent or limit this vasoconstriction may prevent recurrent ischemia and/or myocardial infarction in unstable angina patients. Recently, estrogen receptors were identified in the smooth muscle of post-mortem human coronary arteries. Work in animal models and studies in post-menopausal women suggest that intravenous estrogen acutely decreases coronary vascular resistance, increases coronary blood flow and prevents the paradoxical response to acetylcholine in patients with endothelial dysfunction. This randomized double-blind placebo controlled study tests the hypothesis that the addition of intravenous followed by oral estrogen and of the combination of intravenous and oral estrogen and progesterone to the routine management of unstable angina are beneficial compared with placebo in post-menopausal women. Post-menopausal women with rest angina and no contraindications to hormone therapy will be randomized to receive one of three drug treatment groups, in addition to routine management including intravenous nitroglycerin, aspirin, and heparin. Patients will remain on study drug for three weeks. Initial endpoints during the hospitalization will include ischemia detected on 48 hours of continuous electrocardiographic monitoring, and clinical endpoints of myocardial infarction, death, and need for revascularization. Patients will undergo exercise stress testing and 48 hour monitoring at three weeks and ten days following withdrawal of the study drug.